oral or not ?

[Dominique.Thillaud]

At 14:03 +0200 10/06/97, Birgit Kellner wrote:

>I am quite surprised about what you think I have written. I have nowhere
>expressed such views. Nor have I even claimed to argue for or against
>transmission with or without writing.

	It appears I have actually misunderstanded you.

>My question started from the fact, which is indubitable, that we have
>written texts of considerable length. These texts exist. Some of these
>texts are religious, others are not. Somebody wrote them down at some
>point. Some people claim that these texts have been transmitted orally
>for a long time before they were written down.

	I agree.

>Scholars have assumed
>that a long period of oral transmission leads to a different kind of
>written text than a short one, or no one at all. Oral transmission
>leaves its marks. Therefore, there must be some characteristics in the
>final written text that can tell us something about whether, and about
>how long (in relative terms) the text was transmitted orally before it
>was written down. Some scholars have come up with a number of such
>characteristics. They have done that for a number of texts which have
>nothing to do with India.

	That's right, they have done it. But the problem is the size and
homogeneity of their corpus, both too small. And their analysis of
repetitions, standardized formulas, &c. as mnemotecnics is nothing but an
hypothesis. That can be just a matter of rhetoric: see Shakespeare's
"Honourable men", see RV I, 80 with "arcann anu svarAjyam" and many others.
The true base of their hypothesis is "mnemotecnics are necessary", perhaps
wrong ... (but the story of Truman Capote reported here is a well known
"tour de magie").

>It's not that we have different views, but that we are taking from
>different stages in a process of understanding or research. You appear
>to have already formed an opinion.

	Not already, but until an other one would be more credible by me.
Firstly, I believe using statistics on a text before a stylistic and
pragmatic study is a methodological mistake. Secondly, I'm sure (but
perhaps I'm wrong) that using "cutting and counting" on human productions
is a paranoical behaviour because human productions are never context-free
...

	Best regards,
Dominique

Dominique THILLAUD
Universite' de Nice Sophia-Antipolis, France